Alluvial Gold

Download as pdf or txt
Download as pdf or txt
You are on page 1of 72

0

0
0
0
0
0
0
British Geological Survey
0
0 TECHNICAL REPORT WC/95/5
Overseas Geology Series
0
0 CHARACTERISATION OF GOLD FROM
0
ZIMBABWE:
0
PART 2.
0
0
ALLUVIAL AND SOIL GOLD
M T Styles, P D Wetton and D J Bland
0
0 A Report prepared for the Overseas Development Administration under the ODABGS
0 Technology Development and Research Programme, Project 9211

0
0
0
0 ODA C1assificahon
Subsector: Geoscience
Theme: G1 Promote environmentally sensitive development of non-renewable
0 natural resources
Project Title: Alluvial Gold Characterisationin Exploration Planning
Reference number: R5549
0
Bibliographic reference:
0 M T Styles, P D Wetton and D J Bland
Characterisationof gold from 2imbabwe:part2.
Alluvial and soil gold
0 BGS Technical Repod WCi9515

0 Keywords:
Gold, alluvial, soil, mineralogy, electron microprobe, Zimbabwe.

0 Front cover rik&rahon:


Alluvial gold grains from Zimbabwe
0
0 NERC 1995
0
Keyworth, Nottingham, British Geological Survey, 1995
0
m
0
0
0
0 -

0
EXECUTIVE SUMMARY
e This study of alluvial and soil gold is the second part of a research project on gold in
Zimbabwe, the first part, on bedrock gold was reported by Naden et al(1994). It
0 forms part of a wider study of gold carried out under ODABGS Technology,
0 Development and Research (TDR) project, ‘Alluvial Gold Characterisation in
Exploration Planning’ (Project 92/1, R5549). The TDR Programme is a contribution
a to the British Government programme of aid to developing countries. Zimbabwe is a
major gold producer and the work described here was carried out to test the viability
e of gold characterisation studies in an area of ‘Granite-Greenstone’ - type gold deposits,
a where the geology and mineralisation are more complex than the other areas studied
in Malaysia, Ecuador and Fiji. This study also compares the weathering and alteration
e effects of the semi-arid southern African climatic regime with the tropical rainforest
of Malaysia.
0
Samples of alluvial and soil gold were collected from six areas covering most of the
8 gold producing granite-greenstone terrain of Zimbabwe. Laboratory studies were
.
I carried out to characterise the gold and comparisons made with the bedrock gold from
the same areas and with gold from other countries studied.
e
Studies of size and shape showed that alluvial gold is rounded and equant whereas
e soil gold is angular and irregular. Various size populations were recognised at each
e site but this could not be related to any other features. Morphology indicates whether
alluvial gold is very close to its source but little about the nature of the source.
0
Studies of gold composition showed a wide range of silver contents with many
0 distinct populations reflecting heterogeneous source areas. The dominant silver
content is between is 5- 15% Ag, typical of mesothermal gold deposits. Secondary
gold is only a minor component of alluvial gold, most is similar to the bedrock
0 source.

0 Studies of micro-inclusions in the gold grains show that the inclusions are dominated
by base metal sulphides, pyrite, arsenopyrite and particularly galena. A wide range of
e ‘exotic’minerals were also found, with bismuth and tellurium minerals being most
abundant.
0
8 Alluvial gold clearly inherits many features from its bedrock source in this geological
and climatic regime. The combination of moderate silver contents, with multiple
0 populations, and inclusions of base metal sulphides, bismuth, tellurium, nickel and
cobalt minerals, provides a signature for the granite-greenstone terrain type of gold
0 deposit.
@
0
0
a
0
rn
0
0
0 TABLE OF CONTENTS
0 EXECUTIVE SUMMARY
0 List of Tables
0
List of Figures
0
List of Plates
0
0
1. INTRODUCTION
0
2. SAMPLE COLLECTION
0
0 3. SIZE AND SHAPE STUDIES
3.1 SEM Characterisation
e 3.1.1. Alluvial gold
0 3.1.1.1 Mazoe area
3.1.1.2 Chegutu-Chakari area
0 3.1.1.3 Kadomu area
3.1.1.4 Kwekwe area
m 3.1.1.5 Silobela area
3.1 .I .6 Zvishavane urea
a
0 3.1.2 SoivEluvial gold
3.1.2.1 Mazoe area
0 3.1.2.2 Kadoma area
3.1.2.3 Kwekwe area
0 3.1.2.4 Zvishavane area
a 3.2. Size and shape measurements
a 3.2.1. Alluvial gold
3.2.1.1. Mazoe area
e 3.2.1.2 Chegutu-Chakari area
0 3.2.1.3 Kadoma area
3.2.1.4 Kwekwe area
0 3.2.1.5 Silobela area
3.2.1.6 Zvishavane area
0
3.2.2. SoiVElluvial gold
0
0 3.2.2.1 Mazoe urea
3.2.2.2 Kadoma area
0 3.2.2.3 Kwekwe area
3.2.2.4 Zvishavane area
0
0
0
0
0
0
0 -

e 4. ELECTRON MICROPROBE STUDIES


a 4.1 Mazoe area
4.1.1 Composition of gold
0 4.1.2. Inclusions in gold
0
4.2 Chegutu-Chakari area
0 4.2.1 Composition of gold

0 4.3 Kadoma area


0 4.3.1 Composition of gold
4.3.2. Inclusions in gold
e
0 4.4 Kwekwe area
4.4.1 Composition of gold
0 4.4.2. Inclusions in gold

0 4.5 Silobela area


4.5.1 Composition of gold
0
4.5.2. Inclusions in gold

4.6 Zvishavane area


0 4.6.1 Composition of gold
4.6.2. Inclusions in gold

e
5. DISCUSSION
0
0 6. CONCLUSIONS
@ 7. ACKNOWLEDGEMENTS
0
a - 8. REFERENCES
0 Tables

m Figures
a Plates

0
e
0
e
0
0
0
0
0
List of Tables
0
0 Table 1 List of samples studied.
Table 2 Size and shape of alluvial gold grains from the Mazoe area.
0 Table 3 Size and shape of alluvial gold grains from the Chegutu-Chakari area.
Table 4 Size and shape of alluvial gold grains from the Kadoma area.
0 Table 5 Size and shape of alluvial gold grains from the Kwekwe area.
0 Table 6 Size and shape of alluvial gold grains from the Silobela area.
Table 7 Size and shape of alluvial gold grains from the Zvishavane area.
0 Table 8 Size and shape of soil gold grains from the Mazoe area.
Table 9 Size and shape of soil gold grains from the Kadoma area.
0 Table 10 Size and shape of soil gold grains from the Kwekwe area.
Table 11 Size and shape of soil gold grains from the Zvishavane area.
e Table 12 Electron probe analyses of alluvial gold from the Mazoe area
0 Table 13 Electron probe analyses of soil gold from the Mazoe area
Table 14 Electron probe analyses of alluvial gold from the Chegutu-Chakari
area.
Table 15 Electron probe analyses of alluvial gold from the Kadoma area
0 Table 16 Electron probe analyses of soil gold from the Kadoma area
a Table 17
Table 18
Electron probe analyses of alluvial gold from the Kwekwe area
Electron probe analyses of soil gold from the Kwekwe area
0 Table 19 Electron probe analyses of alluvial gold from the Silobela area
Table 20 Electron probe analyses of alluvial gold from the Zvishavane area
m Table 21 Electron probe analyses of soil gold from the Zvishavane area
Table 22 Electron probe analyses of inclusions in gold from Zimbabwe
0 Table 23 Atomic proportions and identifications of inclusions in gold from
0 Zimbabwe

0
0
a
0
0
a
a
0
0
0
0
0
0
0
0
0
0
0
List of Figures
0
I, Figure 1. Sketch map of Zimbabwe showing the location of areas studied.
Figure 2. Silver content of different samples from ‘C’ mine Zvishavane.
0 Figure 3. S-curves of silver content of alluvial gold from different areas in
Zimbabwe.
0 Figure 4. Inclusion assemblages in alluvial gold.
0
0
0
* List of Plates.

a Plate 1. Photomicrographs of alluvial gold grains from Zimbabwe.


Plate 2. SEM photomicrographs of alluvial gold grains from Zimbabwe.
e Plate 3. SEM photomicrographs of alluvial gold grains from Zimbabwe.
Plate 4.Optical photomicrographs of inclusions in alluvial gold grains from
0 Zimbabwe.
Plate 5. Microchemical maps of alluvial gold grain ZM28/9, showing silver-poor7
e gold-rich tracks, possibly along grain boundaries.
e Plate 6. Microchemical maps of inclusions 4 and 5 in grain ZM28/9 showing complex
BiPbCu sulphide and BiAg tellurosulphide.
m Plate 7. Inclusions 9 and 10 in grain ZM28/9 of galena and complex PbBiTe
sulphide/selenide.
0
0
0
0
0
*
e
0
0

0
0
0
e
0
e
0
0
0
0 -

0 CHARACTERISATION OF ALLUVIAL AND SOIL


GOLD FROM ZIMBABW
0
0 1. INTRODUCTION

0 This report contains the results of work carried out on the ODNBGS Technology
Development and Research (TDR) project 92/1/R5549, entitled 'Alluvial Gold
0 Characterisation in Exploration Planning'. The aim of the project is to characterise
gold grains from bedrock and associated alluvial deposits to establish whether
0 variation in bedrock gold is inherited by the alluvial gold. Further, as this has been
0 demonstrated in other areas, an attempt is made to identify the features found in
alluvial gold which might be used to identify possible source terrains. This
a information can be used to improve the effectiveness of gold exploration by targeting
certain rock types or geological settings at the planning stage of detailed exploration
e programmes. The characterisation covers the study of morphological features such as
a size and shape, chemical composition and mineral inclusions within the gold. An
important feature of the research is the cross correlation between these different
a parameters, such as size and chemical composition, to check for useful relationships

e The work in Zimbabwe was planned to characterise gold from the Archean granite-
greenstone terrain, one of the world's major gold-producing geological settings. It was
a expected that the abundance and detailed variability of gold deposits in Zimbabwe
e would provide good samples to characterise gold and develop the technique.
Collection of gold samples and panning for small amounts of alluvial gold was
0 carried out in collaboration with the Zimbabwe Geological Survey with special
permission from the Ministry of Mines.
0
0 Similar studies are also being carried out in Malaysia, Ecuador and Fiji where
different geological settings and climatic regimes are encountered.
0
0
2. SAMPLE COLLECTION
0
0 Collection of alluvial gold samples during this project was made by hand panning to
produce heavy concentrates at the site of collection. This method enables large
volumes of sediment to be processed on site to give a good indication of the gold
potential (visible gold grains seen in the pan) during sampling. If little gold is found
e at a particular place new sites can be investigated. Collecting large volumes of
e unpanned material was not considered possible as this would involve transporting
large amounts of sample material with no indication of any future value.
0
The serious drought at the time of fieldwork in Zimbabwe meant that only the largest
0 rivers had any water and sample collection could only be done at a few localities
where water as available. This was not necessarily very close to the bedrock sample
0 localities and the collection of closely paired samples of bedrock and proximal
e alluvial gold was only possible in a few places. A close relationship between bedrock
and alluvial gold has been established by work in Malaysia (Henney et al 1994) and
0 this was not considered a crucial shortcoming of this part of the project. Most samples
0
1
0
0
0
0 -
are distal alluvial gold and will show the variety of gold types that are encountered in
0 the catchment areas where there are several possible sources, some of which are
0 represented by the bedrock samples.

0 Samples were collected from ten localities within six areas, shown in Figure 1,
including most of Zimbabwe’s gold-bearing greenstone belts. At least one sample site
0 from each of the areas was studied; the corresponding bedrock gold from these areas
has been described by Naden et al 1994. Location of the samples collected are given
0 in Table 1. Samples of soils overlying gold veins were collected at a few locations.
0 These were mostly from trial pits from recent mineral exploration and samples come
from below the surface horizon from depths around 0.5-2.5m.
0
In the Mazowe area samples were collected from three subareas: a stream close to
0 Oceola Mine that drains the areas containing the mine(ZM 5), the Tategura river that
drains the ground containing Storis Golden Shaft and several other mines (ZM
89),(but was reduced to a few pools at the time of sample collection) and the larger
Mazowe river (ZM90), a few km away downstream of the confluence with the
Tategura, that has many mines in its catchment. These mines are in the central part of
the greenstone belt and are mostly hosted in mafic volcanic rocks.
e Only one alluvial sample was collected in the Chegutu-Chakari area from the
e Mupfure river NW of Chegutu(ZM93). This sample is upstream from the confluence
with the river that has Dalny mine in the catchment and hence no direct comparison
0 with bedrock gold studied during this project can be made.
0 Only one alluvial gold sample was collected in the Kadoma area from the large
Umsweswe river (ZM 85); this drains an area containing many mines that are mostly
c close to the granite-greenstone contact.
0
Alluvial samples were collected from three localities in the Kwekwe area. A sample
0 with abundant gold was collected from the Sebakwe river at Sebakwe Poort (ZM28 8z
29), this drains an area of very productive gold mines, close to the granite-greenstone
contact. A sample with a few grains was collected from the Munyati river at Umniati
0 (ZM35), again draining an area close to the granite-greenstone contact. A further
. sample with abundant gold was collected from the Munyati river near Vein d’Or
0 mine, some 30km downstream from Kwekwe and below the confluence of the
Sebakwe and Munyati rivers (ZM73); this is situated in the centre of the greenstone
a belt. This site is in a river system draining a large area which could have many
possible sources of gold . A supposed soiVelluvial sample from the Grand Parade
9 mine in the Battlefields area was provided by local small-scale miners (ZM36).
0
Alluvial gold samples were collected from one site in the Silobela area on the Gweru
0 river (ZM33 8z 34) which drains the western part of the Kwekwe greenstone belt and
also an area of Karoo (Triassic) and Kalahari (Tertiary) sediments. These younger
0 sediments are not known to be a significant source of gold.
0 A suite of samples were collected in the Zvishavane area from near the ‘C’ mine; this
comprised a bedrock sample and table concentrate described in the previous report,
soil from above the main vein system(ZM68) and an alluvial sample from the
e
0
2
0
0
0
0
Mtshingwe river (ZM69), a few km from the mine. This river drains the western
0 margin of the Mberengwa greenstone belt.
0
The results from each area will be described and compared with the bedrock and soil
0 data.
0
a 3. SIZE AND SHAPE STUDIES
0 The physical morphological features of the gold grains were studied by inspection
with a binocular microscope, detailed surface examination by scanning electron
0 microscope(SEM) and measurements of size and shape using image analysis. The
general morphology of typical alluvial grains from Zimbabwe is shown in Plate 1
where the typical rather rounded form of distal alluvial gold grains can be seen. These
a are from a variety of locations and cover a range of grain sizes.

a 3.1. SEM characterisation.


0
The grains present in each sample were examined by SEM to provide a visual
a morphological description. These observations are presented in a summary below to
give a typical grain description for each sample location.
0
3.1.1. Alluvial gold.
0
0 3.1.1.1. Mazowe area

e Gold grains from this area vary in shape and level of deformation, both between sites
and within each locality. Sample ZM5, sampled in proximity to Oceola mine, shows
a heavily rounded grains with abraded surfaces and folded edge sections. These are a
mixture of elongate and equant nugget shaped grains all showing similar levels of
0 deformation. A typical example is shown in Plate 2a of grain ZM5-21. Undeformed
0 areas on the grains show fine surface structures developed as the gold has formed and
impinged on surrounding mineral grains giving a flattened appearance. The
0 combination of areas of high deformation and areas that show little deformation
indicates the grains are close to the source and have been transported a short distance
0 in a fast flowing stream or river.
e Gold from sample ZM89 shows high levels of deformation with no evidence of
a original surface structure remaining. Grains are varied in shape with equant nuggets
and irregular more elongate examples present. The grains show widespread sub-
0 rounding and folding of protrusions and abrasion of original surface features. Plate
2b of Zh489-8 shows a grain which exhibits two crystal impingement faces and a
e rounded and abraded section. This is a result of the grain recently becoming
a detached from another grain with which it was transported from its source,
undergoing deformation enroute. The surfaces originally in contact with the crystal
e are very smooth at high magnification showing that little deformation of the gold
itself has occurred since the separation.
a
0
0
0 3
0
0 -
Sample ZM90 contains gold that has undergone high-energy deformation. The gold
0 shows extensive sub-rounding and folding superimposed upon mainly nugget shaped
0 grains. This is in contrast to ZM89 where the grains show remnants of complex
original surface structures. Plate 2c and 2d, of grains ZM90-1 and ZM90-1-2
0 respectively, illustrate this with the delicate dendrite like surface clearly visible
indicating the gold has been subject to periods of intense but limited deformation.
0 This may be an indicator of a fast flowing river in which the alluvial deposit has
0 formed a short distance from the original gold source.

0
3.1.1.2. Chegutu-Chakari area
0
Gold sampled from this location shows a range of morphologies consisting of
0 irregular grains of elongate or nugget proportions. In all cases the grains show
0 significant sub-rounding, flattening and folding of protrusions but little evidence of
scratching and abrasion of the grain surfaces. The surfaces show dendritic type
0 structures which show signs of gentle folding and smoothing processes having been
active. This combination of macro scale deformation and preservation of micro scale
0 features indicates the sample location may be relatively close to the original source
and the alluvial environment is low energy and non-aggressive.
0
0
3.1.1.3. Kadoma.
0
Grains from this location show extensive deformation and have rounded, near equant
0 nugget type morphologies. The grain edges are heavily folded and sub-rounded and
0 smoothing of the surfaces by aggressive abrasive impacts is visible. At high
magnification some complex original surface structures are visible within the hollows
0 and folds of the grain surfaces. These suggest a relatively high energy transport to
the alluvial deposit from the initial source but still allowing some features to be
0 preserved.
0
0 3.1.1.4. Kwekwe

0 Gold from ZM28/29 has the form of irregular nuggets and elongated less complex
shapes. The grains have experienced macro scalelhigh energy deformation seen as
0 folding of edges, and flattening of vertices. The gold also shows mineral grains
trapped within the folds as illustrated in Plate 3a of ZM29-6; the white materials
0 embedded in the grain surface are mainly detrital clays plus occasional quartz. High
0 magnification reveals detailed surface structures based upon dendritic growth forms
which have survived the deformation, being protected within the folds, seen in Plate
0 3b. The 'presence of these fine structures and the irregular forms of the folded grains
indicates gold which has not undergone abrasion due to substantial transport from its
0 source to the alluvial deposit. The morphology of gold in sample ZM35 is in marked
0 contrast to this showing much heavier deformation present.

0 The gold of ZM35 shows extensive flattening and deep abrasion of the grain surfaces.
The grains are mainly nuggets with irregular forms some tending towards elongate
0 and show much folding and rounding of their edge sections. Close examination of the
0
4
0
0
0 -
surfaces show depressions on the grains in which dendritic and sponge-like textures
0 survive. These in conjunction with the fresh appearance of the abraded areas suggest
0 the gold is close to its entry point into the alluvial system as smoothing of these
would have occurred if the distance travelled within the alluvial environment was
0 significant.
0 Sample ZM73 contains grains which differ markedly in their morphology. Roughly
half of those examined show similar flattening and heavy deformation to that seen in
0 grains from ZM35 though with more smoothing of their surfaces. The remaining
0 grains show typical nugget forms with folding and rounding of their edges and
smoothing of the grain surfaces. Occasional areas of possibly original surface
0 texture are seen though not on the same scale as in ZM28/29. The presence of two
markedly different morphologies suggests at least two gold sources for the alluvial
0 deposit which would agree with the sample coming from a distal site in a large river
0 where mixing of gold from many sources is possible.

0 3.1.1.5 Silobelu
0 The gold seen at both locations for this area is characterised by surface coatings of
hematite. The coating ranges in form from sub-micron crystals through classic
0 botryoidial hematite to continuous coatings on the gold surface. Gold from ZM33
0 shows more sporadic coatings of the mineral and the grain surfaces show little sign of
high energy deformation having occurred. The grain forms are complex with faceted
e and smooth glazed surfaces where the gold has impinged on surrounding mineral
grains during growth, an example of this is shown in Plate 3f. The plate shows part
0 of a grain displaying a faceted, glazed grain with few scratches or deformation
0 markings visible at approximately 2 , 0 0 0 ~magnification. The smoothed area in the
lower left quarter of the figure is a abrasion smoothed area upon which fine hematite
0 crystals (bright white mass) are beginning to develop.

0 Grains from ZM34 have none of their original surfaces exposed, in all cases being
completely covered by hematite growths. The only exposed gold surface is that due
0 to recent abrasion of the gold within the alluvial system. The overall morphology of
0 the grains is complex with irregular nugget forms common with many protrusions.
. This suggests the gold has suffered little deformation which would have rounded the
0 grains which in turn implies the grains have not travelled far from their source to the
alluvial deposit.
a
The presence of the hematite and other iron oxides found as coatings on the grains is
0 of interest here as at least four differing morphologies are present. The bulk of the
0 grains feature rounded and botryoidal growths which follow the grain surface
intimately Plate 3c. Areas of recently abraded grain surface, where gold has become
0 exposed, 'reveal sub-micron mineral crystals (ZM34- 16-1) which show a definite
iron peak detected by EDX suggesting development of the coating within the alluvial
0 deposit. Plate 3e of ZM34-5-1 shows a grain in which a coating with a lath type
0 morphology has formed on part of the gold grains surface. It is possible that initially
the gold was situated in an iron rich host which may have weathered to form a matrix
0 or surface coating of hematite or iron oxyhydroxides, during this process the lath
structure may have developed. On break up of the host, part of the original coating
0 remained attached to the gold grain as it passed into the alluvial system. Further iron
0
0 5
-
oxide formation has occurred since deposition in the alluvial sediment on top of the
initial coating. An example of the fourth type of coating is shown in Plate 3f of
ZM34-20- 1. Here the coating is a continuous sponge-like structure covering the grain
surface which has within it other mineral matter seen as the bright white masses.

3.1.1.6. Zvishavane

The gold present at this location shows a predominantly equant nugget shaped
morphology with irregular grain forms. The grains show sub-rounding of edges and
protrusions but no evidence of folding, abrasion or plastic deformation which is
characteristic of a high energy alluvial system in which the gold is continually
reworked. The grains show well preserved complex surface structures which display
evidence of gentle smoothing of the features. The grain surfaces suggest a low
transport distance from source to the sample location in a slow moving stream or
river.

3.1.2. SoiYelluviaJ gold

3.1.2.1. Mazowe - samples ZM87, ZM88

Sample ZM87 was collected from the bottom of a trial trench about 2.5m deep in an
area close to Storis Golden Shaft mine. Sample ZM88 was collected from a nearby
area but from a depth of less than 0.5m

The gold present in these samples shows similar morphologies, being dominantly
rounded nugget forms though with a tendency towards elongate proportions in sample
ZM88. Deformation of the gold is limited to scratching and abrasion of the grain
surfaces with folding and sub-rounding rarely evident. Sample ZM88 shows higher
levels of deformation than ZM87 as a result being closer to the surface. The
rounded nature of the gold here and the fine scratching of the surfaces are probably
products of chemical leaching in the soil and possibly some movement within the
soil. Protrusions and irregularities are being gently erased by contact with mineral
. matter within the soil.

-
Kadoma ZM80,ZM82,ZM84

These samples are from soil surrounding a quartz reef and come from an area within a
hundred metres or so of each other. They show differing morphologies across the
three samples. Gold from ZM80 is characterised by very irregular morphologies
which are often clearly angular in nature. Deformation of the gold is minimal in most
cases with the grains showing smooth, glazed surfaces and crystal imprints within
their structures. This type of morphology and surface nature is produced by gold
which has grown within tightly confined mineral veins which have later been
weathered away leaving the crystalline gold forms. Plate 2e shows a typical grain
from this sample with it's angular nature and the impingement of surrounding mineral
grains onto the gold clearly visible. This type of morphology is typical of samples
from soil collected in Malaysia.

6
0
0 L

0 Samples ZM82 and ZM84 contain gold which is significantly different in


0 morphology. The gold from these samples shows a much more rounded form as
illustrated by grain ZM82-2 Plate 2f, where the grains are typified by rounded but
0 irregular nugget forms with little evidence of sub-rounding or abrasion. The surfaces
of the grains are complex in nature with glazed areas present in hollows upon the
0 grain surfaces and show only minor scratching visible at low magnification. The
gold in sample ZM84 shows a higher incidence of this fine scratching but no signs of
0 major deformation. This surface, like that of ZM88 from Mazowe is not typical of
0 soil gold from other areas. The contrasting morphologies within a small area is
noteworthy.

e 3.1.2.3. Kwekwe ZM36

A supposed elluvial sample from the Grand Parade mine in the Battlefields area was
0 provided by local small scale miners (ZM36).

0 Gold from this sample shows evidence of substantial deformation. Many of the
grains are hammered flat, showing edge folds and bending of the main body of the
0 grain. The grain surfaces show coatings of botryoidal hematite which has quartz
e crystals embedded within it. The uncoated grain surfaces show remnants of dendritic
growth structures and irregular features observed at higher magnification. As only
0 six grains were extracted from this sample it is not possible to draw any conclusions
as to the processes which have acted upon the gold or distance from its source. It is
0 possible that the gold was not really elluvial and had actually originated by crushing
a quartz-vein by hand and subsequent hand panning to produce the concentrate. This
activity was being practised by the miners who provided the sample.
0
0 3.1.2.4. Zvishavane. ZM68.
0 The gold in this sample shows a range of morphologies consisting of nuggets and
0 twisted angular elongate grains. The level of macro scale deformation varies with
. some grains showing sub-rounding and abrasion whilst many show little sign of wear.
e The nature of the grain surfaces is varied including grains similar to the alluvial gold
from this area which have complex dendrite based surfaces and grains which show
e angular, faceted surfaces. The morphologies seen are typical of gold near it's source
which has not undergone alluvial transport.
0
0
3.2. Size and shape measurements and analysis.
0
Measurement of the size characteristics of the samples was performed using a
e petrological microscope coupled to a video camera to capture images for input to the
0 image analyser. The samples were mounted on microscope slides and the microscope
operated in transmitted light mode to provide silhouettes of the grains for image
a analysis purposes. The grains were mounted such that their two major axes lay in the
image plane of the microscope so as to present the largest possible cross-section to the
0
0
I)
7
-
analyser. This was done to maximise the validity of data obtained by a two-
dimensional technique operating upon three-dimensional subjects. The image
analysis routine developed features individual grain identification allowing size and
shape data to be cross-referenced to results obtained by EPMA. It is then possible to
explore correlations between sample populations suggested by chemical and
morphological techniques.

The size characterisation was made by measuring area, DMAX, DMIN, perimeter,
FCIRCLE and FSHAPE for each grain in the sample. These are segregated into
primary measurements which are taken directly from the grain and derived
measurements, calculated from the primary measurements.

DMAX and DMIN are the maximum and minimum diameters of grain respectively,
automatically selected from 32 measurements of the grain diameter made at an
angular resolution of 5.7". FSHAPE is a simple aspect ratio for the grain arrived at
by division of the minimum diameter by the maximum diameter of the grain. Objects
with low values of FSHAPE have elongate shapes whilst high values signify objects
which are equant. The function FCIRCLE is a measure of the circularity of the grain,
defined by the equation

where a circle returns the value 1. In addition to indicating the circularity of the
object, FCIRCLE is also a function of the edge roughness. This plotted against
FSHAPE provides a basic morphological characterisation. For example consider a
set of grains where the aspect ratio (FSHAPE) remains approximately constant but
the value of FCIRCLE decreases. This behaviour represents a population where the
general shape of the grains remains the same but the irregularity of their perimeters is
increasing.

The size data obtained were analysed by plotting the data sets in the form of S-curves,
constructed by plotting the data in ascending value order for each of the samples. On
plots of this type related populations of grains tend to plot as a line of constant slope,
while different populations are separated by breaks in slope.

3.2.1. Alluvial gold.

3.2.1.1. Mazowe - ZM5,ZM89, ZM90

Gold from this area shows a low grain size range dominated by fine gold (Table 2).
Grain areas measured lie in the range 0.001 - 0.205mm2 for all but one grain which
has an area of 0.815mm2. The majority of the grains have areas of < 0.1mm2 and
form the first population seen on the S-curves plotted. A second larger grain size
population is indicated for the remaining grains. The grains from each location are
distributed approximately evenly throughout the ranges seen indicating no sorting by
size evident between locations. The shape factor plot for the samples shows a spread
of forms from near equant to strongly elongate morphologies. The grains from

8
-
sample ZM90 predominately fall in the more equant and smoother (circular) zone of
the plot than for the other samples. This suggests that the gold at the two sample
locations may have had a different source with a different original shape and also that
ZM90 has had a longer transport.

3.2.1.2. Chegutu-Chakari - ZM93.

Two clear size distributions are seen in the grains from this area (Table3). These are
0.0125 - 0.0806mm2 and 0.8643 - 2.5053mm2. This distinction is not noticeable on
the shape factor plot where the grains plot as an even spread from the central part
towards the equant and rounded sector of the plot.

-
3.2.1.3. Kadoma ZM85.

This sample shows a wide size range within a single location with definite size
populations (Table 4). The area S-curve shows an overall range of 0.035-0. 1526mm2
with size groupings of CO. lmm2, 0.183-0.623mm2 and >1.0mm2 which are echoed
in the other s-curves. The shape factor plot shows the majority of the grains plotting
in the centre region with a tail towards the equant and rounded region.

3.2.1.4. Kwekwe - ZM28, ZM29, ZM35, ZM73.

These samples show a large range of grain size with areas from 0.010 - 2.814mm2
(Table 5). The lower and upper ends of the range are dominated by grains from
sample ZM28 and ZM73 respectively though grains from all samples are spread
throughout the observed range. The average grain area for the samples was found to
be 0.4mm2 showing the majority of the sample to be of fine to medium gold. The s-
curves plotted for the samples show two size based populations present, these being
grains with areas <0.3mm2 and grains with areas >0.35mm2. The shape factor data
from this area shows the grains to range from equant and smooth in outline to
irregular elongate forms. The majority of grains lie in the centre region of the plot
. having aspect ratios (FSHAPE) and circularity values in the range of 0.4-0.8. This
illustrates the lack of any highly elongate grain forms and the irregularity of the grain
perimeters. The gold shows no shape distinctions from sample to sample with grains
from each site plotting across the entire range present.

-
3.2.1.5. Silobelu ZM33, ZM34.

The gold'from this area is fine grained in nature with a grain area range of 0.012 -
0.111mm2 (Table 6). The S-curves generated show no evidence of differing size
populations being present. The grains from ZM33 can be seen to be smaller than
those in ZM34 from the area s-curve but show no other differing characteristics. This
might be a result of slightly different sediment sources and panning methods by the
assistants who collected the samples. The shape factor plot shows the grain form to
be intimately mixed across the samples. The majority of grains have aspect ratios
<0.5 indicating the dominant morphology to be nugget based with the wide range in

9
0
e
0 -
circularity parameter (0.307- 0.751) for these showing the variation in complexity of
a the grain perimeters. Three of the grains sampled stand out from the rest having
0 circularity and aspect ratios below 0.4 indicating highly elongate morphologies.

0
-
3.2.1.6 Zvishavane ZM69
0
The gold from this location displays two grain size populations within the size range
0 seen (Table 7). The first of these is 0.0122-0.0988mm2, the second being 0.2021-
e 0.6635mm2 with the grains split evenly between the two ranges. The shape factor
plot for the grains shows the grains to be mainly near equant nugget forms with a
0 tendency towards irregular perimeters.
a
a 3.2.2. SoivElluvial Gold.

e Only a small amount of the fieldwork time was spent collecting soil gold. The
samples were washed, seived to less than 2mm and hand panned in the field to
a remove clay, organic material and most light minerals. The resulting concentrate was
then processed with a ‘Gold Genie’ spiral concentrator.
0
a -
3.2.2.1. Mazowe ZM87,ZM88.
0
Shape data for these samples shows the gold to be fine grained in nature with two
0 probable size populations (Table 8). The total size range for the samples is 0.0094 -
0.1036mm2 with the majority of the grains falling in the range 0.0094 - 0.0256mm2.
The shape factor data shows a distribution with a tightly constrained scatter of points
a (excluding grain ZM88.2). The aspect ratio range is small for the distribution whilst
circularity varies more widely indicating grains with similar basic geometry but
a complex and differing perimeter morphology.
e
0 -
3.2.2.2. Kadoma ZM80, ZM82, ZM84.

0 The shape data for the Kadoma elluvial samples shows little evidence of differing
shape or size populations present (Table 9). The gold is fine grained having a size
a range of 0.0029-0.0520mm2~across which grains from all three samples are spread,
and average grain size of 0.019mm2. The shape factor plot shows a wide distribution
0 of grains ranging from rounded, equant forms into elongate and rough or irregular
0 forms. This wide scatter of points on the shape plot is more typical of alluvial gold
than soil gold, which in other areas show a more restricted range. This allied to the
e rounded surfaces observed by SEM could possibly imply a component of alluvial
gold in the soil.
a
0
a
0
0
0 10
0
0
0 -

a -
3.2.2.3. Kwekwe ZM36.

a The small number of grains in this sample make meaningful analysis of the results
impossible, sizes range from 0.0089- 0.11 14mm2 indicating a fine grain size. The
0 results are given in Table 10.

0
8 -
3.2.2.4. Zvishavane ZM68.

0 This sample shows a fine overall grain size having the range 0.065-0.0346mm2
(Table 11). This is split into two size ranges above and below approximately
0.017mm2 on the area S-curve. This is not apparent in the other size data plots or in
the shape factor plot for the sample. The shape factor data shows the grains to have
e limited ranges for aspect ratio but a large range in circularity. This shows a dominant
0 general geometry, in most cases mildly elongate grains, which have a widely varying
outline morphology. The small range of aspect ratios seen when considered with the
a low deformation seen for many of these grains is probably representative of the range
of gold morphology within the source deposit
0
0
0 4. ELECTRON MICROPROBE STUDIES
0 4.1. Mazowe area
a 4.1.1. Composition of gold.
0
The results of the electron probe analyses of all the alluvial samples from the area are
a shown in Table 12. The only element consistently detected in amounts significantly
above detection limits is Ag and the variation in Ag is plotted as a modified S curve.
a On this type of plot different of compositional populations are distinguished by
e breaks of slope in the curve. The data from Mazowe shows several populations, one
with very low Ag, less than 2wt%, a large population with some variation between 5-
0 10% Ag, probably another group with 10-15 Ag% and a distinct high-Ag group with
greater than 20% Ag.
a
a The gold from the bedrock samples from the mines each showed very narrow ranges
0 of composition with mean Ag contents of 3 and 5.3 % Ag. The samples of table
concentrate however showed much larger ranges, but with the vast majority in the
0 range 3-10%, a few between 10-1596 and a few with greater than 30% Ag (electrum).
0 The soil samples collected near Storis Golden Shaft yielded only a few gold grains,
0 analyses of which are shown in Table 13. Most compositions were between 2-5%
Ag, similar to the bedrock gold. One grain had 23.5% Ag, contents this high were not
0 seen in the bedrock samples but were found in the table concentrates.
0 In this area the alluvial gold was collected relatively close to the localities of bedrock
gold (less than 1Okm) and the compositional populations distinguished in the alluvial
0
0
0 11
-
gold can mostly be identified in the bedrock. Both sample types have the majority of
compositions with between 510% Ag, with some with lower Ag, which could be the
lower end of this bedrock composition population. Some grains with less than 1% Ag
are probably of secondary origin. Plate 4e shows a grain from ZM90 with abundant
pure secondary gold. Secondary gold generally occurs as a narrow rim on the gold but
appears to be particularly abundant on this grain as the section through this grain
happens to be close to one end of the grain. The smaller population with 10-15% Ag
and those rich in Ag were also found in the table concentrate and the latter also in the
soil.

In this area the alluvial gold, which is relatively close to its source, records a range
and complexity of Ag contents that have been identified from studies of bedrock and
overburden gold.

4.1.2. Inclusions in gold.

Inclusions in alluvial gold were analysed by electron microprobe and the results are
given in Table 22. These inclusions are often very small and atomic proportions and
identifications are given in Table 23.

The inclusions found in the gold from the Mazowe area were mostly base metal
sulphides, pyrite, arsenopyrite and galena, which are also the common sulphide
minerals in the gold veins in the mines in the area. One grain of tetradymite, Bi2Te2S,
was found, similar to that found in the table concentrate from the nearby Oceola
mine. Also two grains of possible acanthite-argentite, A&, were found. Argentite is
often associated with 'epithermal' mineralisation and this along with some of the
silver-rich gold could indicate that some gold may be related to a later, lower-
temperature phase of mineralisation. Plate 4a shows pyrite in an alluvial grain from
ZM5 and plate 5d shows coarser pyrite inclusions in a table concentrate from Oceola
mine.

4.2. Chegutu - Chakari area


4.2.1. Composition of the gold

The results of electron microprobe analyses of alluvial gold grains are given in Table
14, along with a modified S-curve of silver contents. The S-curve is complex, with
similar populations of silver content to that found in other areas. These are low-Ag
(<5% Ag), moderate-Ag(5-10 % Ag) and high-Ag (10-20% Ag), with a few very
pure gold.grains. These compositional populations bear no relationship to the size
populations identified.

The bedrock samples from Dalny mine were silver-rich around 1520% Ag with
some nearer electrum (30% Ag). This would match the Ag-rich compositions found
in the alluvial samples. Those with lower silver are similar to the dominant
compositions found in other greenstone belt areas, containing 5- 15% Ag and might be
expected here.

12
0
e
0
0 No inclusion were found in the gold from this area. This is unusual and has only
0 rarely been encountered in studies of many other areas world-wide.

0
4.3. Kadoma
0
0 4.3.1. Composition of gold

0 The alluvial gold grain compositions are given in Table 15, along with a modified S-
curve of the silver compositions. Most grains in this area had detectable mercury, up
0 to 0.8%, whereas in many areas mercury is below detection. This indicates that there
e is some regional variation in mercury content of gold. Silver compositions show
several populations, a small low-Ag group with less than 2% Ag, the vast majority in
a a population with 5 1 5 % Ag and some high-Ag grains with more than 20% Ag. There
is no apparent relationship to the size populations identified.

The soil samples are all from a small area of trial pitting within a hundred metres or
0 so of each other and can be considered as one sample site. The results of analysis are
0 given in Table 16, along with a modified S-curve for silver. The results show a very
similar pattern to that of the alluvial gold, the bulk of the samples in the range 5- 15%
0 Ag with a few low and a few high values. The range of silver compositions found in
soil and alluvial gold is virtually the same, which suggests that this range is a feature
0 of the bedrock mineralisation andor in situ weathering processes rather than a feature
0 formed during transport.

0 The majority of compositions found in alluvial gold, in the 515% range, is the same
as that found in bedrock samples and reflects the main gold mineralisation. Some of
a the low-Ag values are from the edges of grains and are probably due to, secondary
processes but others could be due to internal inhomogeneities (see next section). The
0 very high-Ag values, greater than 20%, have not been found in bedrock samples but
0 as these are relatively small in number this could be that there are not enough bedrock
samples to cover the whole range of compositions. Alternatively this might be a
0 supergene effect, but a process that produces Ag-rich gold in the supergene
environment has not, to our knowledge, ever been described or suggested in the
0 literature. Known supergene processes tend to form very pure gold, as Ag-rich gold is
more soluble and tends to dissolve. Pure gold may be reprecipitated, with silver
e removed in solution.
0
0 4.3.2. Inclusions in gold

Inclusions in alluvial gold were analysed by electron microprobe and the results are
a given in Table 22. These inclusions are often very small and atomic proportions and
identifications are given in Table 23.
0
The inclusion types identified were dominated by base metal sulphides, pyrite galena,
chalcopyrite and other copper sulphides. These base metals are common in the gold
mineral veins. The more exotic minerals found in some of the bedrock samples were

0
0 13
0
e
e -

a not found in the alluvial gold. This was however only a small alluvial sample, only
20 grains, and a larger number of grains is necessary to improve the chances of
0 obtaining a more representative suite of inclusions.

a
4.4. Kwekwe
a
a 4.4.1. Composition of gold

a Electron microprobe analyses of alluvial gold are given in Table 17, along with
modified S-curves showing the variation in Ag content. Small amounts of other
e elements were found , particularly quite high levels of Hg, around 3%. This is much
a higher than alluvial gold from other areas studied in Zimbabwe but was also found in
bedrock gold from Indarama mine within the catchment area. This again indicates
a regional variation in gold composition. The silver S-curve shows a complex pattern,
with a large population with very low Ag, up to about 3%, another large population
a with 5 1 0 % Ag, a further smaller group with 10-20% Ag and one electrum with 31%
Ag. All the grains from the Munyati sample (ZM35) fall in the 5 1 0 % group, the
a Sebakwe Poort samples dominate the low-Ag group but also include a few high-Ag
e grains and the most distal sample, ZM73 covers the whole range

0 The analyses of the elluvial sample are given in Table 18, along with an S-curve of
Ag contents. This shows a population with 5 1 0 % Ag similar to the nearby alluvial
a sample, ZM35 and a couple of electrums with 30% Ag. The range of compositions is
e similar to that found in the alluvial samples but there is no evidence in this soil
sample of the formation of very pure(>l% Ag) secondary gold.
e
The populations seen in the alluvial gold can be related to compositions found in
a bedrock samples, the low Ag group(>5% Ag) are very similar to compositions from
Gaika and Indarama mines, the latter also having high Hg. The paint gold, from ores
0 from the near surface oxidised zone, from the Broomstock extension mine near
0 Indarama and Ivanhoe mine also had very low Ag gold, similar to many alluvial
grains. This type of gold is probably formed by secondary processes and is
e - indistinguishable from gold formed by other secondary processes. Indarama mine also
had a minor component of Ag-rich gold with 20-25% Ag, similar to some alluvial
0 grains. The table concentrate from Goldleaf mine, also in the catchment area showed
a range from 5 3 0 % Ag, possibly from one population and also an aurian silver with
0 55% Ag. The wide range of Ag contents found in alluvial gold can be matched with
e composition from known bedrock sources within the area.

e Microchemical maps were made of one grain from the Sebakwe Poort sample,( the
method for making microchemical maps was described by Styles et al 1993). The
maps for gold and silver Plate 5 , show that most of the grain has a fairly uniform
a silver content, which quantitative analysis shows to be around 11%. There are
however gold-rich tracks, bright red on the gold map, which contain less than 1% Ag
e and appear to form along subgrain boundaries. This is a feature that has been seen in
samples from many areas but the origin is not known, possibly they form by some
a kind of exsolution process.
a
e
0 14
0
a
e
0 4.4.2. Inclusions in gold.
0
Inclusions in alluvial gold were analysed by electron microprobe and the results are
e given in Table 22. These inclusions are often very small and atomic proportions and
identifications are given in Table 23. Inclusions were very abundant in this area, with
more inclusions found in the Sebakwe Poort samples than all the other samples from
0 Zimbabwe added together.

0 The inclusion assemblages are again dominated by base metal sulphides, pyrite
arsenopyrite and galena, with sphalerite also found in this area. Several rare minerals
e were found, with the most abundant being tetradymite (Bi2Te2S),with lesser
cobaltite/glaucodot and complex PbBi sulphide minerals and minor hessite (Ag2Te),
a chalcostibite and stibnite. The antimony minerals are widespread in the area and were
e found in bedrock samples from Gaika and Indarama mines. These antimony minerals
are a characteristic of the Kwekwe area and it is perhaps a little surprising that they
e did not form a larger proportion of the inclusions. Tetradymite and glaucodot have
been found at several places in Zimbabwe, in the greenstone belt gold mines, but not
e in the bedrock samples from the Kwekwe area. There are many gold mines that drain
e into these rivers and only a very small proportion were sampled and hence those
studied here will not represent all possible sources.
0
The nature of some of these inclusions and particularly those rich in silver was not
0 clear from the point analyses, and microchemical maps were made in an attempt to
clarify the situation. The maps shown in Plate 6 show an area about 50mm square in
Q grain ZM29/8, covering inclusions ZM29/8 4 and 5 from Tables 22 and 23. Two
0 sulphide inclusions are clearly shown on the S map. The upper grain contains Pb, Bi
and Cu while the lower grain contains Te, Bi and Ag. The silver bearing BiTeS
0 mineral is situated along one of the silver poor tracks that show clearly on the Ag
map. The tracks may be formed by exsolution and silver exsolved from the gold may
0 have combined with a tetradmyite like mineral. The general composition of these
minerals can be seen from the maps but recalculating the point analysis to a specific
e mineral formula is difficult, but the grains have approximate formulae CuPbBi2S9and
Q Ag3Bi2Te3S2respectively. These formulae have not been matched to previously
described minerals.
e
Maps of inclusions ZM29/8 9 and 10 from Tables 22 and 23 also show two sulphide
e inclusions shown in Figure 5 . The upper inclusion (10) is the lead sulphide galena
e but also appears to contain a small amount of selenium. The lower one (9) contains
many elements, Pb, Bi, Te, Se and S, while the lack of a 'hole' in the Ag map shows
0 it also contains Ag. To a resolution of about a micron all the elements appear to be in
a single phase but such a mixture cannot be matched to a single known mineral and
0 the possibility of a very fine mixture must be considered possible.
0
e
0
0
a
0 15
0
0
e
4.5. Silobela
e
0 4.5.1. Composition of gold

0 Electron microprobe analyses of alluvial gold are given in Table 19, along with
modified S-curves showing the variation in Ag content. The S-curve shows a simpler
a pattern than in most other areas in Zimbabwe with a very dominant population with
0 10-17% Ag. There are minor groups around 5% Ag, 20% Ag and two grains greater
than 30% Ag. The gold grains consistently contain around 0.2% Hg, which is
0 probably natural, but a few rims have much higher Hg (10%) which is the result of
contamination by mercury used by alluvial 'miners' working in the area. Plate 4c
e shows a grain that has been affected by mercury where the amalgam has been leached
out to leave very porous gold with some parts with high Hg. Blobs of metallic
0 mercury were seen in the pans during sample collection in this area.
0
The Ag-content populations show derivation from a source with a main
a mineralisation with gold containing around 15% Ag, but with minor components
containing both higher and lower Ag. The Ag content of the few bedrock samples
a collected is very low, as these were from near surface oxidised zone deposits, where
there is low-Ag, secondary gold. The main population in the alluvial samples is
however typical of many other deep mine samples from greenstone belt deposits
0 elsewhere in Zimbabwe. The two size populations show no consistent differences in
composition and may be the result of different sample collection methods.
0
e 4.5.2. Inclusions in Gold.
e Inclusions in alluvial gold were analysed by electron microprobe and the results are
a given in Table 22. These inclusions are often very small and atomic proportions and
identifications are given in Table 23.
0
0 Very few inclusions were found in the gold in this area but they showed considerable
variety. They were galena, hessite (Ag2Te), lillianite (PbBi)2S3, tetradymite (Bi2Te2S)
0 and cinnabar, possibly with galenobismuthite(PbBi)S. The bedrock samples from this
area were very few and gold was hand picked fro m vein quartz and little is known
0 about other minerals. Galena and tellurides are known from the other areas studied in
the Zimbabwe greenstone belts.
0
e
0 4.6. Zvishavane
0 4.6.1. Composition of the gold
0 The results of electron microprobe analysis of the alluvial gold along with a modified
0 S-curve of silver contents is given in Table 20. This shows a dominant population
with 5-15% Ag, possibly with a minor division around 10%. There is also a group
a with high-Ag around 30-40% Ag and one low-Ag grain.
0
0
e 16
0
0
e -
The analyses of gold from the soil are given in Table 21, along with an S-curve for
a silver content. This shows a main population with 20-3096 Ag a minor group around
0 10% and one low-Ag grain.

0 The bedrock sample, with coarse visible gold contained around 10% Ag but the table
concentrate from ore with fine-grained gold showed a dominant composition around
0 15% Ag with minor components around 10% and 20%. The soil gold therefore had a
e minor component around 10% Ag similar to that from deep underground samples but
the majority were higher in silver than the underground samples. This could indicate
0 that there is some vertical zonation within the mine, with Ag-rich gold near the
surface. The alluvial sample contains a mixture of the components from soil and rock
0 with the dominant component the main underground mineralisation with 10-15% Ag.
The 3'-curves for all these sample types, rock, table concentrate soil and alluvium are
e shown together in Fig 2 to show the similarities between the sample types.
e
a 4.6.2. Inclusions in gold.

0 Very few gold grains from this area contained inclusions and the results are given in
Tables 22 and 23. The inclusions were all galena apart from one iron oxide that
a presumably is secondary in origin. The number of inclusions is too small to be
0 confident that they are representative and the presence essentially of only one type of
inclusion is unusual.
0
0
0 5. DISCUSSION

0 The size and shape studies have produced a large amount of information about
alluvial gold in Zimbabwe. SEM examination of alluvial gold clearly shows the
0 effects of rounding and abrasion suffered by the gold grains, which is much greater
0 than that seen in proximal alluvial gold in Malaysia. There is however often fine
surface detail preserved in the hollows on these grains. Some grains from Silobela
0 showed clear evidence of secondary overgrowths of iron oxides/hydroxides. The
contrast between the rounded nature of the alluvial grains and the angular nature of
0 gold from soil was most marked, the one exception being the rounded gold from
e some soil samples in the Kadoma area.

e The size and shape measurements show that the gold studied was generally fine-
grained, most grains having a maximum diameter less than lmm. Many samples
0 contained several size populations but these could not be related to other features.
The alluvial samples all show similar shape characteristics on the FCIRCLEFSHAPE
e diagram. Values of both functions show a proportional increase from values around
0 0.4 to 0.8 leading to a cluster of points in the centre of the diagrams with a variable
trend towards a diagonal distribution. This range of shapes indicates a trend of
a increasing sphericity with increasing smoothness, as would be expected from the
modification during alluvial transport. The soil gold in contrast shows a very different
e pattern with a narrow range of FCIRCLE values and a spread of FSHAPE giving a
0 horizontal distribution of points. This implies a change in circularity or roughness but

a
17
0
e
e
little change in aspect ratio. The exception again to this is the soil gold from Kadoma
e which has a similar range of shape characteristics to the alluvial gold. Possibly there
0 is some alluvial gold in the soil in this area.

0 The size and shape studies generally confirm and quantify the features that can be
seen from observation with the binocular microscope. They also have the drawback
a that they are only two dimensional and do not show the generally flattened nature of
e many of the grains, which give the appearance of being equant on the plots.
Measurements of this type, based on the silhouettes of grains, do not differentiate
0 between a sphere and a disc, as both appear as a circle in projection.
e The electron microprobe analyses of the composition of the alluvial gold grains
showed a broadly similar range of compositions from all the areas studied. This is
a shown in Figure 3. The silver content shows the most variation and the majority have
a compositions in a range 5 1 5 % Ag with minor components with higher and lower
contents. The range of compositions is similar to that found in bedrock gold in the
e same areas. There are often several subgroups within the 5 1 5 % range which would
be in accord with the narrow range of compositions found at single mines but
e reflecting the variation from mine to mine even within one area. Gold from the
e Kwekwe area has a large population of pure gold. Some of the very pure gold may be
secondary gold such as 'paint gold' from oxidised zones formed due to secondary
e processes which is widespread in this area. The presence of 'pure' gold tracks,
possibly along subgrain boundaries, was found in several samples and will account
e for some of these analyses. Some of the silver-rich gold may be due to a lower
temperature 'epithermal' overprint. The wide range of compositions and the multiple
e populations found in most areas is shown in Figure 3. This wide range seems to be a
0 feature of greenstone belt type deposits and contrast to the simpler patterns found in
the mesothermal, slate belt type deposits studied in Malaysia, where much simpler
e patterns found. The dominant population in the Zimbabwe samples 5 1 5 % Ag is
however similar to Malaysia and characteristic of mesothermal deposits.
0
Mercury shows some variation from area to area being high in the Kwekwe area and
0 low in Chegutu and Zvishavane areas. This presumably reflects variations in
e mineralising fluids.

e The soil gold show a similar range of compositions to the bedrock gold in most areas
but several contain high-Ag gold, in some cases with higher Ag contents than that
e seen in the bedrock gold. In the soil from 'C' mine near Zvishavane this high-Ag gold
e was the dominant component. This is surprising as Ag-rich gold is generally thought
to be less stable and more readily dissolved in the surface environment and a
e predominance of low silver gold might be expected.
e The study of inclusions in alluvial gold showed the dominance of base metal
sulphides, pyrite, arsenopyrite, galena and in some places minor pyrrhotite and
a chalcopyrite. The proportions of these minerals varied considerably from area to area
e but galena was always abundant. A bar chart showing the inclusion assemblages from
the different areas is shown in Figure 4.Many areas show a significant proportion of
e bismuth and bismuth lead minerals and tellurides. Similar minerals were found in
Ecuador and one area in Malaysia, the common theme is a proximity to granitic
0 intrusions. A further distinctive mineral group was the cobaltite-glaucodot
0
0 18
e
0
e I

association, found in some areas. This has been seen elsewhere associated with basic
0 rocks. The characteristic feature of the granite-greenstone terrain could be the
0 combination of all of these types of inclusions, base metal sulphides, bismuth
tellurides and cobaltite.
e
0 Alluvial gold characterisation probably has most potential in poorly known areas
a where large areas are being considered for gold exploration. In this situation the
alluvial gold supplies information about possible gold targets from a wide area. The
0 technique is possibly not at its most effective in a country like Zimbabwe where gold
deposits are plentiful and relatively well known. As stated in the introduction the
0 main aims of the work here were to test the characterisation methods in the semi-arid
climate and to collect information on the greenstone type of gold deposit. These
e objectives have been accomplished and several others possibilities for application in
0 this setting are considered.

0 Alluvial gold studies in Zimbabwe have revealed wide ranges in silver contents and
inclusion assemblages from sample sites where there are many possible sources.
e Sampling with more restricted catchment areas could be used to check if known
sources can be matched with those identified in alluvium or if further, as yet
0 unknown, sources are indicated. In other areas types of gold mineralisation, different
0 to the standard greenstone type are indicated, such as ‘epithermal’ types. Checks for
and examination of alluvial gold outside the main greenstone area could be made as it
0 is possible that all gold is assumed to come from the greenstone belts and other types
might be overlooked. This should be considered as different exploration strategy
0 might be required. If further studies are carried out, more detailed investigations of
0 the mineralogy of bedrock gold are necessary to fully characterise known deposits.
Our studies of many mines found accessory minerals not mentioned in previous
0 studies and a good knowledge of these is vital for comparison with inclusions in
alluvial gold.
0
Gold characterisation studies could be carried out in Zimbabwe as there are electron
e microprobes at both the Institute of Mining Research and the University of
0 Zimbabwe. After appropriate training in both the rationale of the approach to gold
. exploration and training in the specific techniques, these methods could be used in
e Zimbabwe, possibly in association with local mining companies.
e
0 6. CONCLUSIONS

0 1. Size and shape studies show that alluvial gold has a different shape from soil gold
as is apparent from inspection with a binocular microscope and confirmed by
0 quantitative measurement. Soil gold is very angular and irregular while alluvial
gold is rounded and either equant or flattened . This rounding occurs after short
e transport distances. SEM studies show far more detail about surface features than
e optical examination, giving information about transport history and mineral
overgrowths. Several size populations were recognised from all sites.
0
0
e
19
0
0
0
a -
2. Studies of the composition of gold show that there is a clear regional variation in
0 the mercury content of gold, this being particularly high in the Kwekwe area.
0 Alluvial gold from all areas shows a wide range of silver contents, comprising
several distinct populations. The dominant compositional group is 5- 15% Ag.
e These compositional groups cannot be related to the size populations. Soil gold
shows a similar range of compositions to bedrock gold but in some places with an
a unusual additional component of silver-rich gold. The origin of this silver-rich
0 gold is not known, but could be a low-temperature overprint on the main
mesothermal mineralisation.
0
3. Very low silver, secondary gold, is a minor component of alluvial gold, except
a possibly in the Kwekwe area. It is also minor in the soil samples studied, though
elsewhere it is a major component of near surface ores from the oxidised
e supergene zone. Much of the alluvial gold is identical to primary bedrock gold in
e this climatic and weathering regime.

a 4. Studies of the micro-inclusions in gold showed that the assemblages of inclusions


are dominated by the base metal sulphides pyrite, arsenopyrite and galena but
0 with proportions varying from area to area. Bismuth and tellurium minerals are
relatively abundant in several areas. A wide range of rare complex sulphide and
a telluride minerals, some not definitely identified, were also found
a 5. Many features of distal alluvial gold can be matched with features known from
a bedrock gold in the catchment area. This confirms that alluvial gold inherits
features from its source, as was shown from previous studies of proximal gold.
e
e 6. The signature of gold deposits occurring in granite-greenstone belt terrains appears
to be those associated with mesothermal shear zones; moderate silver contents and
a inclusions of base metal sulphides, with the additional presence of a wide range of
types of inclusions, including both those associated with granites, (bismuth
a minerals and tellurides) and those associated with mafic rocks, (cobaltite and
gersdorffite). The wide range and multiple populations of silver content in alluvial
0 gold, reflecting local variation in bedrock gold within small areas, may also be
0 characteristic.

e 7. This work has demonstrated that alluvial gold characterisation studies could be
used effectively in the semi-arid tropical climatic regime, as abundant primary
e gold is present in alluvium. Widespread secondary gold would easily be
recognised but has not been found. A granite-greenstone signature could be
a recognised in large scale reconnaissance surveys. The detailed complexity of the
0 greenstone belt gold deposits means that very extensive sampling and detailed
studies would be required to locate individual sources within the deposits.
a
a rl
1. ACKNOWLEDGEMENTS
a
We would like to thank Dr P J Pitfield and Dr C D G Campbell of the ODA-BGS TC
0 Midlands goldfields project for advice and assistance with the fieldwork and
e
0
a 20
0
0
0 -
discussion of results. The Geological Survey of Zimbabwe is thanked for assistance
a with arrangements and provision of field assistants and equipment.
a
a
7.REFERENCES.
a
Naden, J, Styles, M T and Henney, P J. 1994. Characterisation of bedrock gold from
0 Zimbabwe. British Geological Survey Technical Report WC/94/5 1
0
Henney, P.J., Styles, M.T., Bland, D.J. and Wetton P.D. 1994. Characterisation of
a gold from the Lubuk Mandi area, Terengganu, Malaysia. British Geological Survey
Technical Report WC/94/2 1.
a
0 Styles, M.T., Perez-Alvarez, M,. Bland, D.J., Wetton P and Naden J. 1993.
Characterisation of gold from Ecuador. British Geological Survey Technical Report
a WG/92/48.

a
0
a
a
0
0
a
a
a
a
e
e
a
0
e
0
a
0
0
0
a 21
0
0 -

0 Tables
0
0
0
0
0
0
0
0
0
0
0
0
0
0
a
0
0
a
0
e
e
0
0
m
0
0
0
0
0
0
L

TABLE 1. LOCATIONS OF ALLUVIAL AND SOIL GOLD SAMPLES.

Sample Sample
Number Type Locality Grid Ref

MAZOWE AREA
ZM 5 [Alluvial Murowadzi river nr Oceola Mine, Mazowe
ZM 87 Soil Pits nr Storis Golden Shaft Mine, Mazowe 2 836 80 625
n
ZM 88 Soil 2 834 80623
ZM 89 Alluvial Tategura river, Mazowe 2 861 80 628
ZM 90 Alluvial Mazowe river, Mazowe 2890 80692

KADOMA AREA
ZM 80 ]Soil Heroine Mine area, Kadoma 7 970 79 626
II
ZM 82 ]Soil
I
ZM 84 Soil
ZM 85 Alluvial Umswezi river, Kadoma

KWEKWE AREA
ZM 28 lAlluvial Sebakwe River, Sebakwe Poort, Kwekwe 7 900 79 137
ZM 29 \Alluvial 1"
ZM 35 IAlluvial IMunvati River, Umniati
ZM 36 Elluvial? Grand Parade Mine, Umniati 7 963 79 382
ZM 73 Alluvial Munyati River, nr Empress mine 7 6 9 0 59 577
I

ISILOBELA
I I

AREA 1

ZM 33 Alluvial Gweru River, Silobela 7 253 79 159


ZM 34 Alluvial

IZVISHAVANE
I I I
AREA I I

ZM 68 Isoil IAbove vein nr 'C' mine, Belingwe, Zvishavane 17 985 77 326


ZM 69 1 Alluvial I Mtshinawe river. Belinawe. Zvishavane 18 020 77 400
Table 2. Shape measurements for Mazowe Gold

1. ZM5 0 ZM89 + ZM90 I 1.6 ..


0.8
zE 0.6
. c.

f
..
c.

2
-
$
rn
Q)
0.4
i!
0.8

0.4
0

0.0
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Count Count

. 1 .o
0
0.
.8
9 I 0.8
0.6
0
O O
.
**

2
3 0.4

0.2

0.0 h ~ 0.0
0 10 20 30 40 50 60 70 0.0 0.2 0.4 0.6 0.8 1.0
Count FCIRCLE
0
a Table 3. Shape measurements for Chegutu AI luvial Gold
0
a
0
0

0
0
0
e
e
e
0
0
..
...
3.0 2.5
0
2.0
0 EE2-5 ifl
0
2.0
1.5
-E 1.5
0
c
I

1.0
2= 1.0
..
P
a 0.5 0.5
m
..m
....
0 0.0 1 I

0 5 10 15 20 25
0
count
m
0 1.4 1 .o IRouah & I /Rounded
0 1.2

.
0.8
~

1.0
E 0.8 8! 0.6
e 0.6 9
p
s 0.4

0
0 0.4
0.2 Elongate I
0 0.0
0.0 0.2 0.4 0.6 0.8 1.0
0 FCIRCLE

0
Table 4

T
T
..=I E
2.5
..
. E 1.5 ..
.
c.

.. ...
..mmmm.
0.0 1 0.0 J 1

0 5 10 15 20 0 5 10 15 20

I
I
Count
1
1
Count

I 1 1 - --I

r .. .. .
..
...;......
1.2
0.8
1.0
E. 0.8
c ..= h
9
2 0.4
0.6
I

0.0 1 0.0 1 1

0 5 10 15 20 0.0 0.2 0.4 0.6 0.8 1.0


Count FCIRCLE
I I I

Page I
Table 5. Shape measurements for Kwekwe Alluvial Gold
0
0
I ame 3. anape measuremenis Tor nweKwe Aiiuviai uoia
e
0
0 1 ZM28 0 ZM29 + ZM35 ZM73 2.5 0

0 8
3.0 2.0
0
p
2.5

2.0
0

EE
U
1.5
d
0 c.
1.5 @ 2B 1.o
n
e
0
e 0.0 4
0 20 40 60 80
1

0 20 40 60 80
0 Count Count

e
e 1.2 1.o

0 0.8
0.9
e EE 0.6
0 U
0.6 a
#
0 I
n
E 0.4 D o

0.2

0
0.0 4 1 0.0
0 0 20 40 60 80 0.0 0.2 0.4 0.6 0.8 1.0
Count FCIRCLE
0
0
0
0
0
0
0
0
0
0
0
0
a
a Table 6 .Shape measurements for Silobela Alluvial Gold
a
0
0
0
a
1.522 I 0.604 I 0.663
ZM33.10 0.040 0.334 0.217 1.134 0.394 0.649 1.024 0.606 0.553
0 ZM33.11
2M33.12
0.063
0.029
0.357
0.268
0.255
0.172
1.189
1.004
0.559
0.356
0.716
0.641
1.483
2.101
0.511
0.285
0.507
0.345
0.978 0.514 0.523
0
1.557 0.336 0.579
0 1.437 0.392 0.549

a
a 12111133.211
~ 0.012 I 0.174 I 0.103 I 0.513 I 0.564 I 0.592
.__-

a 2111133.22
2133.23
0.028
0.018
0.276
0.235
I
I
0.167
0.116
0.837
0.635
0.494
0.552
0.605
0.492
1.292
1.378
0.634
0.462
0.663
0.645
e 2133.24 0.019 0.201 I 0.134 0.620 0.619 0.664 1.174 0.470 0.698
2133.25 0.016 0.195 I 0.123 0.648 0.469 0.633 0.902 0.697 0761
12133.261 0.014 I 0.183 I 0.127 I 0.608 I 0.472 1- 06G-
0
a 0.1 2 2.0
a 0.10
1.8
1.6
E #- 1.4
0 0.08
E 1.2
a -9m 0.06 P O

2
U
0.04 n 0.6
0 0.02

0 0.00 I 0.0 J
0 20 40 60 0 20 40 60
0 Count Count
0
a 0.4 1 .o

0 0.3 0.8

0 EE
a
c.

L
z
0.2
5
8
O O 6

0.4
Tl

a
0.1

0.0 I
0.2

0.0
.
0 20 40 60
a count
a
a
0
0
Table 7. Shape measurements for Zvishavane Alluvial gold
0
0
0
0
0
0
1ZM69.101 0.2323 I 0.8748 I 0.4226 I 2.4434 I 0.4889 I 0.4831 I
0
0
0
0
0
0
0
0 0.6
0-7 1 . 1.2 1 ..=
n

g 1.0
....
1
0 E 0.5 --
E
0.4 -. # UE 0.8
.
0
0 1
0.3 --
2 0.2 -. .1
8
2 0.6
zp 0.4 .fl.

a
0
Oel
0.0 L
0 5 10 15 20 25
0.2
0.0
0 5 10 15 20 25
1

Count
0
0
0 . 1.0 ,El y Equant

-
.
0.7 0.8 -.
0
0.5
1
h
a
0.6 _-
0 0.4 1 1
0.4 --
0
0
0 ~
0.3
0.2
0.1
0.0
0 5 10 15 20 25
Oe2
0.0
IF]:
0.0 0.2 0.4 0.6 0.8 1.0
; ,

Count FCIRCLE
0
0
0
0
0
0
0
a -
Table 8. Size and shape of soil gold from the Mazoe area
a
0

0
0
a
a
a
a
a

0
0
- 1 0-70
0.60 0

0
0 m
0
E 0.40 J 8

0
0.00 1
0 0 5 10 15 20 25 0 5 10 15 20 25
Count
0 Count

a
1.o

- 0.25
Oe30 I
I
.U

0.8
IEquant I I & Equant
h 0.6
9 0.4 U

0.05 ' 0.2


0.0
0.0 0.2 0.4 0.6 0.8 1.0
FCIRCLE
Table 9.Shape measurements for Kadoma Soil Gold
c

8 ZM80 0 ZM82 ZM84


0.35 .

0.080 E 0.25 a.

0.060
c .*&@

1
4***
0.040 P

0 10 20 30 40
0.05
0'0° 0
i 10 20 30 40
Count count

0.20 0
.
8

-
8
E
i
#
0.15

0.10
4m*

$ 0.4 1
0.05
Elongate
0.00 1 0.0
0 10 20 30 40 0.0 0.2 0.4 0.6 0.8 1.0
count FCIRCLE
0
-
Table 10. Shape measurements of Kwekwe soil gold.
NAME AREA DMAX DMlN PERlM FCIRCLE FSHAPE
Zm36.1 0.1114 0.4835 0.3476 1.6296 0.5271 0.7189
0 Zm36.2 0.0253 0.2210 0.1537 0.7764 0.5271 0.6955
Zm363 ~ 0.0089 0.1281 0.0899 0.3921 0.7277 0.7018
zm36.4 0.0151 0.1967 0.1447 0.7617 0.3264 0.7356
Zm36.5 0.0179 0.2347 0.1205 0.6484 0.5350 0.5134
Zm36.6 0.0118 0.1615 0.1154 0.5283 0.5328 0.7146 L

0.120
0.100
rn
0.450 --
.
0.400 -.
a E 0.080 0.350 -.
E 0.300 - -
$
c. 0.060 -.
m
2 0.040
U
0.020
....
n 0.150 -.
0.100 -.
.
O.oO0 i
I
o.oO0 I
. 0.050
0 2 4 6 8 0 2 4 6 8
Count Count
a
0.350
0.300 1 . 1.o
0.9
0.8

-E 0.200 t
0.250

.
0.7
h
0.150 .. 2
0.6
0.5
0.4
0.3
0.2
O.oO0
0.050 i
0
I I
2 4 6 8
0.1
0.0
0.0
Elongate

0.2 0.4 0.6 0.8 1.0


Count FCIRCLE

0
0 Table 11. Shiape measurements for Zvishavane Soil Gold
0

0
a
0
0
0
0
*
a
0
0
0 0.035 - .= T .
0 E 0.030 ..
0.025 ..
. 8
0.35
0.30 I..

0
6
UI 0.020

U
(0
0.015 ..
2 0.010 ..
a 0.005 ..
.’

. ....-
..
#mu..

o.OOo-. : 0-05
0.00 i
0 5 10 15 20 25 30
Count

0.2

0.2
1.o
0.8
rzYl
Equant

-E 0.1
.. 8!
9
0.6

E 0.1 I
p 0.4
0.2
0.0 0.0
0 5 10 15 20 25 30 0.0 0.2 0.4 0.6 0.8 1.0
Count FCIRCLE
0
0
0 -
0 Table 12. Electron microprobe analyses of alluvial gold from the Mazoe area.

0
0
0
0 zm5/6 0.00 0.00 6.89 0.00 0.09 0.04 0.00 91.76 0.00 98.78
zm517 0.08 0.00 4.79 0.00 0.00 0.00 0.06 93.77 0.19 98.88
zm518 0.11 0.00 19.88 0.10 0.00 0.00 0.00 80.50 0.27 100.86
0 zm519 I
0.00 1
0.00 8.42
1 1
0.00 1
0.00 1
0.00 1
0.06 91.35
1
0.21 1
100.04
lzm5/10 I 0.001 0.001 6.381 0.001 0.001 0.001 0.061 93.821 0.101 100.361
0 zm5111 0.08 0.00 7.32 0.00 0.00 0.00 0.07 92.19 0.23 99.88
zm5112 0.00 0.09 19.80 0.00 0.00 0.00 0.00 81.08 0.25 101.21
0 zm5113 0.00 0.07 6.65 0.00 0.09 0.00 0.08 93.78 0.22 100.89

0
0
0 zm5/19 I 0.001 0.001 6.88 0.00 0.001 0.00 0.041 92.851 0.00 99.77
0 zm5/21 I 0.001 0.001 6.61 0.00 0.001 0.00 0.081 92.811 0.21 99.70

0
0
0
e
0
0
Silver 'S' Curve
0
a

1 1
25/ U

a ***
0 ZM90
15
.*
0 ZM5 ***
10
0 e * =

0 1o l P = . * * * * * * * *
5
:
0
1

0 5 10 15 20 25 30
0 Number

0
0
0
0
0
0
0
e Table 13. Electron microprobe analyses of soil gold from the Mazoe area.

As S Ag Te Fe cu Au Hg Total
a zm8714
zm8716
0.00
0.00
0.12
0.00
2.82
4.91
0.00
0.10
0.00
0.00
0.00
0.00
94.63
94.84
1.64
0.67
99.21
100.53
0 zm8717 0.00 0.00 4.05 0.00 0.00 0.17 95.63 0.37 100.22
zm8718 0.00 0.00 2.81 0.00 0.00 0.10 96.01 0.61 99.53
0 zm8814 0.00 0.00 2.69 0.00 0.03 1 0.001 96.37 0.16 99.26
zm8816 0.00 0.00 5.34 , 0.09 0.001 0.181 94.71 0.12 100.44
0
1

zm8716 0.14 0.00 5.09 0.00 0.00 0.05 94.56 0.65 100.48
zm8718 0.00 0.00 2.48 0.10 0.00 0.10 96.99 0.62 100.29
0 zm8816 0.08 0.00 4.78 0.00 0.00 0.16 94.65 0.10 99.77
zm88/7 0.1 1 I 0.051 23.521 0.11 1 0.06 I 0.001 74.621 1.27 I 99.74
a
0
0 'S' Curve for Silver

e
e $
ae
30
20
10
.
0
0
0
0

0
0

0
0
0
a
0
0
0
0
e
e Table 14. Electron microprobe analyses of alluvial gold from the Chegutu area.
0 As S Ag Te Fe 'CU Au Hg Total
zm9311 0.09 0.00 2.41 0.10 0.00 0.00 97.53 0.18 100.32
0 zm9312 0.09 0.08 12.36 0.00 0.00 0.00 89.23 0.00 101.75

a
0
0
0

e
a
e
e
0
e
0
e
a
0
0
0
e
'S' Curve for Silver
e
0
25

.....
... .=
20
0 15
arn .==
0 s 10

0
0
Number
0
0
0
0
0
0
0
e Table 15. Electron microprobe analyses of alluvial gold from Kadoma area.
0 As S Ag Te Fe cu AU Hg Total
zm8511 0.00 0.16 0.38 0.09 0.00 0.00 97.99 0.19 98.81
0 0.00 74.36 0.00 102.47
zm8512 0.00 0.07 28.03 0.00 0.00
a zm85/3 1 0.00 0.00 9.18 0.00 0.00 0.00 90.16 0.12 99.46
zm8514 j 0.00 0.00 10.52 0.00 0.00 0.04 89.06 0.25 99.86
0 zm85/5 i 0.00 0.00 12.88 0.12 0.00 0.00 87.43 0.18 100.62

0
e
a zm85/14a 0.00 0.00 5.22 0.00 0.03 0.00 95.21 0.30 100.76
0 zm85/14b 0.08 0.00 5.17 0.00 0.04 0.00 94.87 0.29 100.45
zm85/16 0.00 0.00 0.51 0.00 0.00 0.05 98.96 1
0.00 99.52
_ _ __
zm85/17a 0.00 0.00 8.13 0.00 0.00 0.07 92.061 0.00 100.26
zm85/17b 0.00 0.00 7.97 0.00 0.00 0.06 92.731 0.00 100.76
0 zm85/19a 1 0.07 0.00 5.80 0.10 0.00 0.00 92.16 0.11 98.24
zm85/19b ~ 0.00 0.00 0.00 0.09 0.93 0.00 88.92 0.00 89.95
0 zm85/20 1 0.00 0.00 9.53 0.00 0.00 0.05 90.49 0.30 100.37

0
0
Silver 'S' Curve

30.00
0 25.00
20.00
0 3 15.00
0 10.00
5.00 ...mm
0 0.00 m m

0 5 10 15 20 25
0
j Number

a
a
a
0
0
0
0
0
0

e
c Table 16. Electron microprobe analyes of soil gold from the Kadoma area.
0 As 5 Ag Te Fe cu Au Hg Total
zm8018 0.12 0.00 8.72 0.00 0.00 0.00 91.70 0.00 100.54
0 zm82/1 0.00 0.00 8.70 0.00 0.04 0.00 91.58 0.22 100.54

0
0
0
0
a
0
0
e
0
e
0

'S' Curve for Silver


a 25
0 20 8

8
0 U
15
8 10
0
5
0 8 8 .

20
e 0 5 10
Number
15 25

0
0
a
0
0
0

0
e
0
0
'able 17. Electron microprobe analyses of alluvial gold from the Kwek:Wre are
a
e
0
0
a
0
e
0
a
a
a
a

0
0
0
a
m
0
0
a
e
a
e
a
a
m
a
0
a
0
0
0

Table 17. Electron microprobe analyses of alluvial gold from the Kwekwe area
-

a lzm29/4 I 0.001
. . 0.001I 2.381
1
0.101 0.001 0.001 97.11 1 0.001 99.591
zm29/5 0.00 0.001 8.111 0.00 0.001 0.00 91.651 0.151 99.90
0 zm2916
~ ~~ ~ 0.14 0.001 7.491 0.00 0.001 0.06 91.951 0.471 100.11
zm29/7 0.00 0.00 11.59 0.09 0.00 0.001 88.501 0.521 100.70
a zm29/8 , 0.00 0.00 11.61 0.00 0.00 0.101 89.301 0.001 101.01
zm29/9 0.07 0.00 1.36 0.00 0.00 0.10 96.28 1.78 99.58
0 zm29110 0.00 0.00 2.23 0.00 0.03 0.00 97.59 0.21 100.05
zm29112 , 0.00 0.00 0.61 0.00 0.00 0.00 99.45 0.13 100.19
0 zm29113 1 0.11 0.00
1

14.681 0.00 0.03 0.00 85.371 0.28 100.48


zm29/14 0.00 0.00 8.261 0.00 0.00 0.00 91.751 0.16 100.18
0
~

0
e
a
a

a
a
0
a
a
0
0
0
a
a Silver 'S'Curve

a 40.00
35.00 ZM35
a 30.00
0 ZM28l29
25.00
0
0
. ':20.00
15.00
10.00
ZM73

5.00
0 0.00
0 0 20 40 60 80 100
Number
0
0
e
Table 18. Electron microprobe analyses of soil gold from the Kwekwe area.
e
0 As S Ag Sb Te Fe Cu Au Hg Total
zm36/1a 0.00 0.00 34.07 0.00 0.10 0.00 2.14 66.67 0.00 102.98
0 zm36/lb 0.08 0.00 31.41 0.00 0.00 0.00 0.75 69.18 0.00 101.43

0
0
0
a
e 'S' Curve for Silver
a
35.00 m
e 30.00
25.00
e g 20.00
8 15.00
e 10.00
m . '
5 .OO
0
0
0
0
0
0
0
e
0
e
0
e
0
0
a

0
0
0
0

Table 19. Electron microprobe analyses of alluvial gold from the Silobela
0 -
area.
I Se I As I S I Ag Sb I Te I Fe 1 cu 1 Au I Hg I Total
zm33/a I 0.001 0.071 0.001 4.561 0.001 0.001 0.041 1.831 80.551 9.861 96.92
0 zm33/b
i

I 0.00 0.08 0.09 12.771 0.00 0.00 0.001 0.001 86.951 0.371 100.25
zm33/1 1 0.00 0.00 0.00 14.131 0.00 0.10 0.001 0.001 86.421 0.251 100.90
0

0
0
0
e
0
0
0
e
0
0
0
0
0
0

0
0
0
zm34/21 j 0.001 0.00 0.35 9.74 0.00 0.00 0.00 1.03 61.32 27.26 99.70
0 zm34/22 1 0.001 0.00 0.00 16.59 0.00 0.00 0.00 0.00 82.36 0.00 98.94

0
0
a
Table 19. Electron microprobe analyses of alluvial gold from the Silobela
area. -

lzm33128 I 0.001 0.001 0.001 14.321 0.001 0.001 0.001 0.001 84.131 0.331 98.771

"S" Curve for Silver

50
40
..
30
aa
ae 20
10

Ob
0
... 10 20 30 40 50 60
Number
Table 20. Electron microprobe analyses of alluvial gold from the Zvishavane area.

AS S Ag Sb Te Fe CO AU Hg Total
zm69/1 0.00 0.00 7.12 0.00 0.00 0.00 0.00 92.23 0.10 99.45
zm69/2 0.00 0.00 6.55 0.09 0.00 0.03 0.05 93.14 0.21 100.07
0.00 0.00
a zm69/3
zm69/4
0.00
0.00
0.00
0.001
35.75
18.47
0.00
0.00
0.08
0.091 0.001 0.051
66.84
82.461
0.15
0.321
102.82
101.39
a zm69/5
zm6916a
0.16
0.00
0.001
0.00
13.40
8.09
0.00 ,
0.00
0.001
0.00
0.001
0.00
0.001
0.00
87.361
92.53
0.001
0.17
100.921
100.79
zm6916b 0.00 0.00 8.40 0.00 0.00 0.03 0.05 92.72 0.15 101.34
zm69/7 I
0.00 I
0.00 I
5.43 I 0.00 I
0.00 , 0.00 0.00 95.57 0.00 101.01
tzm69/8 I 0.001 0.001 14.21) 0.001 0.001 0.001 0.001 86.351 0.501 101.061
zm69/9 0.09 0.001 14.15 O.OO( 0.00 0.001 0.091 86.851 0.001 101.18
a zm69/10 I
0.12 0.001 12.01 0.001 0.00 0.041 I
0.061 89.271 0.111 101.62
zm69/11 0.00 0.06 3.56 0.00 0.00 0.16 0.00 96.00 0.00 99.78
e zm69/15 0.13 0.00 30.80 0.09 0.00 0.00 0.00 71.00 0.00 102.02
zm69/16 0.00 0.00 5.33 0.00 0.00 0.03 0.00 95.50 0.11 100.97
a zm69/20 0.10 0.00 15.80 0.00 0.00 0.00 0.00 85.70 0.00 101.60

a lzm69/9
zm69/10
I 0.101
0.00
0.001 13.881
0.00 11.58
0.001
0.00
0.001
0.00
0.001
0.00
0.101
0.00
86.401
88.53
0.001
0.00
100.471
100.11
a zm69/11
zm69/12
0.00
0.00
0.00 1.85
0.00 10.19
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
96.74
90.33
0.00
0.00
98.59
100.52
zm69/14 1 0.001 0.001 7.73 0.00 0.00 0.001 0.00 92.75 0.00 100.48
zm69/15 1 0.121 0.001 31.00 0.00 0.00 0.041 0.00 70.74 0.00 101.90
. _ _
zm69/16 0.00 0.00 9.12 0.00 0.10 0.00 0.00 91.01 0.00 100.23
zm69/19 0.12 0.00 6.40 0.00 0.00 0.00 0.08 93.44 0.40 100.44
zm69/20 I 0.00 0.00 15.99 0.00 0.00 0.00 0.00 84.93 0.09 101.00

a
'S' Curve for Silver
a

m
40

25
....
' t
a- 2o
15

0 5 10 15 20 25 30
Number

a
0

0
0 Table 21. Electron microprobe analyses of soil gold from 'C' mine Zvishavane

0
e
e
0
0
0
a
0
0
a
0
0
" S Curve for Silver
I'

0
40
35

. .. . .
.
30
0
0
25
20
15
10 .... .
0 5 , 9
04
0 5 10 15 20
Number
0
a
8
*
a
0
0
0
0

0
0
0
0
0
a
0
a
0
0
0
0
e
0
a
e
0
e
e
0
0
e
a

a
0
0
0

0
a
e
0
0
-
:?a?
Y O ( D
r m

-_-

--
;
n
0
0 0
83 82
mm
0

Q .-cm
n E
s!?
m m
E sE ms e .E .E
D % %
i n n
i(i rag i o 0
I-OP ---
L O O
E ??-: ? ? ?
3 0 0
0 - 0 0
L
rc
___
2 ? ? ?
c oo
? ? ?
3 0 0
0 m
rn
___
.-E "99
0 0 0
? N ?
> o o
U)
C
___
.-0 22: ? ? ?
-
U)
3
0
NLDm
3 - 0
3 m *

_--
.-
C ?N?
m
0
o o
? ? ?
> o o
y..
0
___
U) c?? ? ? ?
I= > o o
0
.I
*(DO
*
c,
Q
.-
0 28: ? ? ?
> o o
!k
c,
c ? ? ? ? ? ?
a 0 0 0 ) O O
E
U ? ? ?
C ? ? ?
0
0 0 0 > o o
U)
C
.-0 288 !??
> o o
tl
0
n
2
n
997
0
olo
.-0 9-9
0 0 0
E
0
? ? ?
2 0 0 0

4
cv c??
0 0 0

a
5
z 7 9 9

? ? ?

!
? Y ??a!
t j rg(val
m N

m m
z .E
% %
P P
0 0
3 I
. C C a
tm tm h.E

? > ? ? ?
0 0 0

? ?
>
828
? ? NY?
I 0 0 0

? ? p!??
m m m
> > *mcD

! ! ???
3 0 0

! ? ???
3 0 0

! ! ???
3 0 0

! ! ggg
! ? ? ?
I 3 0 0

I
! ? ? ? ! ? ? ?
I 3 0 0 I000

I
I ? ? ?
I 1000

, ,999
, , 0 0 0

! S"?
- 0 0

? ? 91
0 0 0

7???l
0 0 0
0
1
310/ 0 1010

I
? ? ? ? ?
3 0 0 0 0

ygzg
? O O O O
310 0 0 0
D
D
B
Acanthite
D Tetradymite
Galena
Mazowe
Arsenopyrite
Pyrite
0 10 20 30
0
0
-
0 Galena
I
, -

0 Arsenopyrite Kadoma
Pyrite
0 I

b 20 40 60
0
a
Others
e Hessite
a Coba Itite
e Bi-minerals Kwekwe
Galena
0 Arsenopyrite

-
0 Pyrite
0 5 10 15 20
0
-
0 Hessite
0 1-1
Bi-minerals
Silobela
Galena
0
0
e
Arsenopyrite
Pyrite
L
0 10 20 30 40

0
a 1 -
Galena T
0 Arsenopyrite -t Zvishavane
Pyrite
0 * I

0 50 100
0
0
Figure 4. Inclusion assemblages in alluvial
0
gold from Zimbabwe
0
0
-
Figures
c

Figure 1. Simplified geological map of Zimbabwe


showing the locations of the areas studied. Map
modified from Foster 1989.
0
0 -
0 Figure 2. Ag content of different samples from 'C' mine
0
'S' Curve for Silver

a
1: 1
0
0
0
0
1 d

-
...................................
20.00

.'"!I 10

TABLE CONCENTRATE
Number
20 30

a I
'S' Curve for Silver
0
* I 0
40
30 I
I
0
0
0 20 40 60
a Number

0
SOIL
0
I I

a
0
0 dl
...... ..
0
0
0
e ALLUVIAL

0 'S' Curve for Silver

a 40 1

0
0 10
0 .........................
0 0 10 20 30
Number
0
0
0
0
0
0 -
Figure 3. S-curves of silver content of alluvial gold from different areas in Zimbabwe.
0
0
0
a 1 .. ..
... .
...
20
25

0
0
2 15.
a9 1 0 .
5
''
...
.........
. , ! 5 ,

5
,

0
0 .L 0
0 10 20 30 0 10 20 30 40 50 60 70
0 Number Number

Mazoe Kwekwe

0
20
25 1 1
1 50
40 I
0
5 /
0 0
0 0 20 40 0 20 40 60
0 Number Number

0
Chegutu-Chakari Silobela
0
0
0 25
I

t .
8

20
25 I c
1

0 0
4
15
s

1
10
5
e 0
-O 0 20 40
0
Number Number
a
0 Kadoma Zvishavane

0
0
0
0
0
I)

-
Plates
0

0
e

e
e
0
e

0
0
0
0
0
e

0
0
e
e
e
0
0
0
0
PLATE 2

0
0
0
0
0
0
0
0 2a. Alluvial gold grain from sample 2b. Alluvial gold grain from sample
ZM5/21 showing typical rounded form ZM89/8 showing angular form and
0 negative crystal shape.
0
0
0
0
0
0
0
0
0
2c. Alluvial gold grain from sample 2d. Detail of ZM90/1 showing fine
0 ZM90/1 showing rounded form and structures in depressions between
0 abraded surfaces. Scale bar 1Ovm. abraded areas.
0
0
0
0
0
0
0
0
0
0
0
PLATE 3

3a. Alluvial gold from sample ZM29/6


showing rounded form. 3b. Surface detail of grain ZM29/6.

3c. Alluvial gold grain ZM 3412 showing 3d. Alluvial gold grain ZM34/20 showing
fine Fe-rich overgrowths . porus overgrowths.

3e. Alluvial gold grain ZM34/5 showing 3f. Alluvial gold grain ZM33/15 showing
fat h-like overgrowths. clean faceted surfaces
I

You might also like